Users can create GIF animations in a single step using a text prompt or an existing image, and edit them frame by frame afterward.
According to OpenAI, the feature is powered by the new Images 2.6 model, which generates images 50% faster than its predecessor. The model comes in two versions: GPT-Image-2.6 Sunburst and GPT-Image-2.6 Flare. The first delivers higher-quality image generation, while the second can generate an image in just six seconds.
ChatGPT could already create GIF files, but the process previously required users to submit multiple prompts. With the new update, animations can now be generated with a single prompt.
Users do not need to provide a highly detailed image, either. Even a simple sketch is enough to create an animation. After generation, users can adjust the animation speed and edit individual frames directly in the chatbot.
OpenAI has not yet specified the maximum animation length, resolution, or usage limits.
The update is available to all ChatGPT and Codex users on the web, as well as through desktop and mobile apps.
Additionally, the image annotation feature already available in ChatGPT has now been added to the API.
